SUPPORT, UNDERMINING, AND NEWCOMER ......make the process of adjustment especially difficult. Another unresolved issue for socialization re-search involves time. Socialization is a

SUPPORT, UNDERMINING, AND NEWCOMERSOCIALIZATION: FITTING IN DURING THE FIRST 90 DAYS

JOHN KAMMEYER-MUELLERCONNIE WANBERG

University of Minnesota

ALEX RUBENSTEINUniversity of Florida

ZHAOLI SONGNational University of Singapore

While much organizational socialization occurs through interpersonal interactions,evidence regarding how these processes unfold over time has not been forthcoming.Results from a 14-wave longitudinal study with a sample of 264 organizational new-comers show that support of newcomers from coworkers and supervisors declineswithin the first 90 days of employment. Early support and undermining had moresignificant relationships with work outcomes assessed after 90 days of employmentthan did increases or decreases in support and undermining over that time period,suggesting early support and undermining may lay a foundation for later work out-comes. Proactive behavior partially mediated the relationship between support andmore distal work outcomes, including withdrawal behaviors. Supervisor underminingwas uniquely associated with higher turnover (exit) hazard.

When most people think back to their first dayson a new job, they are prone to recall positive andnegative social interactions. Some will remember aparticularly tense conversation with a supervisorthat made them realize that their adjustmentwould not be easy, whereas others remember espe-cially helpful colleagues who steered them in theright direction. But are there really significant im-pacts from these early interpersonal interactions?The academic literature on newcomer socializationhas been comparatively silent on this front. Thepreponderance of the socialization literature hasplaced either the broader organization or the new-comer as the key drivers of socialization. Meta-analytic summaries of the literature either focus onorganizational tactics alone as the exogenoussource of socialization (Saks, Uggerslev, & Fassina,2007), or only include newcomer information seek-ing and organizational tactics (Bauer, Bodner, Er-dogan, Truxillo, & Tucker, 2007). In general, co-workers and supervisors have been entirelysubsumed under the “organization” side of social-

ization, or otherwise are simply treated as a passivesource of information available only after newcom-ers initiate information seeking via proactivity.Clearly, however, insiders can make an effort tohelp newcomers adjust or even, in certain cases,make the process of adjustment especially difficult.

Another unresolved issue for socialization re-search involves time. Socialization is a dynamicprocess, involving change and evolution (Fisher,1986). Some researchers have proposed that initialinteractions can have powerful effects on subse-quent outcomes (e.g., Wanous, 1992), highlightingthe need to measure interactions, behaviors, andattitudes very soon after organizational entry todetermine how they have effects further down theline. Others propose that changes over time areimportant (Chen, Ployhart, Cooper-Thomas, Ander-son, & Bliese, 2011), so it is necessary to measurenot just the initial status of newcomers, but also theways in which processes play out over time withineach individual. At present, the question of howdynamic processes of socialization play out in re-lation to one another is unanswered.

In a study incorporating both socialization dy-namics and interpersonal processes, Jokisaari andNurmi (2009) illustrated that among new employ-

Thank you to Mo Wang for his helpful advice on adraft of this work.

1104

� Academy of Management Journal2013, Vol. 56, No. 4, 1104–1124.http://dx.doi.org/10.5465/amj.2010.0791

Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s expresswritten permission. Users may print, download, or email articles for individual use only.

ees, declines in supervisor support were related todecreases in newcomer role clarity and job satisfac-tion, and a slower rate of salary growth. We extendtheory about the role of interpersonal processes insocialization in three substantial ways. First, weexamine the role of not only supervisor support,but also coworker support. Comparing supervisorand coworker relationships allows us to see hownewcomers might differentially respond to interac-tions with different agents of socialization (Kam-meyer-Mueller & Wanberg, 2003). Second, re-sponding to a call by Bauer et al. (2007), we explorethe extent to which destructive behaviors (co-worker and supervisor undermining) hinder new-comer adjustment. Finally, we examine whetherpositive hedonic tone (i.e., positive mood, happi-ness, or pleasant feelings at work) and proactivesocialization behaviors (i.e., volitional activities toassist adaptation to the workplace) explain howand through what mechanisms support and under-mining impact newcomer adjustment outcomes.This approach complements a growing body of re-search demonstrating the importance of moods andemotions in organizational behavior (e.g., Elfen-bein, 2008; Fisher & Ashkanasy, 2000; Weiss &Cropanzano, 1996) as well as the emphasis on new-comer proactivity (e.g., Ashford & Black, 1996;Wanberg & Kammeyer-Mueller, 2000).

To investigate these relationships in a way that isconsistent with our emphasis on the temporal na-ture of the socialization process, we used a repeat-ed-measures design, surveying newcomers weeklyover a period of three months. We incorporate alatent growth modeling strategy (e.g., Chan &Schmitt, 2000; Jokisaari & Nurmi, 2009; Lance,Vandenberg, & Self, 2000), allowing us to observethe independent effects of the initial status of vari-ables (e.g., at start of employment) and changes inthose variables (e.g., slopes or trajectories) overtime. This strategy enables us to demonstrate theimportance of social support in the first 90 days ofnew employment, extending previous research,and also to test hypotheses regarding changes insocial undermining that have not been examined.

THEORETICAL DEVELOPMENT ANDHYPOTHESES

The conceptual model guiding our study isshown in Figure 1. We begin with a discussion ofthe role supervisor and coworker support and un-dermining are posited to play for newcomers overtime using a combination of affective events theory(AET) and socialization theories. We then describehow newcomer hedonic tone and proactive social-ization will mediate those processes. Hedonic tone

FIGURE 1Model of Social Support, Undermining, Hedonic Tone, Proactive Socialization, and

Newcomer Adjustment Outcomes

SupervisorSupport

SupervisorUndermining

CoworkerSupport

CoworkerUndermining

HedonicTone

ProactiveSocialization

VoluntaryTurnover

WithdrawalBehaviors

OrganizationalCommitment

SocialIntegration

WorkProactivity

Intercept Slope

Intercept Slope

Intercept SlopeIntercept Slope

Intercept Slope

2013 1105Kammeyer-Mueller, Wanberg, Rubenstein, and Song

represents one pole of an affect “circumplex” withhappiness and pleasant feelings on one end andunpleasantness and sadness on the other end (e.g.,Elfenbein, 2008; Russell, 1980; Weiss & Cropan-zano, 1996). We focused on hedonic tone as a high-er-order index of affect rather than assessing posi-tive and negative affect separately. Though lesscommon in organizational research than the posi-tive/negative affectivity (PA/NA) model, a hedonictone model confers notable theoretical and meth-odological advantages. First, at the within-personlevel of analysis, positive and negative affect arestrongly negatively correlated with one another(Scollon, Diener, Oishi, & Biswas-Diener, 2005),making state (temporary) mood under the PA/NAmodel difficult to interpret. To that end, the PA/NAmodel might be more useful as a trait-based affec-tivity model; however, for the purposes of ourweekly study time frame, hedonic tone may bettercapture changes in mood (Weiss & Cropanzano,1996). Second, research suggests that hedonic tonemodels are more robust in factor analysis acrossdifferent forms of surveying affect (Russell & Ridge-way, 1983) and also show better fit to data (Green,Goldman, & Salovey, 1993). Third, hedonic tone isparticularly important for organizational newcom-ers, since it is likely to be the initial affective ap-praisal made as employees react to work events(Frijda, 1986).

According to AET (Weiss & Cropanzano, 1996),events influence affective experiences at work. In-dividuals will first be exposed to some incomingstimulus and will base their choice about con-sciously attending to it on its inherent novelty orworth. Following this, they will appraise the situ-ation relative to their self-schemata, to learn if theincoming stimulus contains some sort of relevantmeaning and has the potential to impact them(Plutchik, 1994). The more positive (or negative) anevent, and the more personally important the eventis to the person, the stronger the emotional reactionthat is likely to be experienced (Carver & Scheier,1998). Employees report that work events related tointerpersonal interactions produce some of themost salient emotional responses (Boudens, 2005;Elfenbein, 2008; Kelly & Barsade, 2001). In a qual-itative study, Boudens (2005) found that some ofthe most frequently cited positive triggers includepersonal support and connectedness, while nega-tive triggers included interpersonal conflict, powerstruggles, and violations of trust.

In terms of the AET appraisal process, comparedto more tenured employees, during their first weeks

on a job newcomers are likely to consider most, ifnot all, of their work experiences to be novel andmeaningful, since each new piece of informationobtained on the job facilitates the process of makingsense of an unfamiliar environment (Lewin, 1951;Weick, 1995). As such, events such as interpersonalinteractions are likely to resonate especiallystrongly with newcomers, thus establishing thefoundation for future attitudes and job behaviors.In the present study, we focus on social supportand social undermining as positive and negativepatterns of interpersonal events that each affect theadjustment process. Further, recognizing that “in-terpersonal interaction” is somewhat of an um-brella term, we aimed to sharpen our conceptualunderstanding by exploring how two differentsources of support and undermining (i.e., cowork-ers and supervisors) uniquely contribute tosocialization.

Theoretical work in the area of organizationalsocialization has proposed that established organi-zational insiders can exert especially strong influ-ences on newcomer adjustment, filling in whereorganizational socialization efforts fail to provideinformation and social validation (Moreland &Levine, 2001). These organizational insiders repre-sent the social setting into which a newcomer isattempting to fit (Miller & Jablin, 1991), so signs ofsocial support can be interpreted as evidence thatthe environment for the newcomer will be positiveand accepting, whereas undermining will send astrong signal that the environment for the new-comer will be negative and rejecting. We discussthese issues in greater depth now.

Support, Undermining, and Hedonic Tone

Social support consists of “transactions with oth-ers that provide the target person (i.e., the recipi-ent) with emotional support, affirmation of the self,appraisal of the situation, instrumental support,and information” (Vinokur & Van Ryn, 1993: 350).Organizational newcomers may be particularly inneed of social support because of the uncertaintythat accompanies meeting new people and learningnew tasks. Supervisors and coworkers are in anideal position to provide newcomers with assis-tance in adjusting to the demands of their work role(Moreland & Levine, 2001). We examine both co-worker and supervisor support and underminingboth because prior research shows these twosources have complementary effects on employeeattitudes and behaviors (Chiaburu & Harrison,

1106 AugustAcademy of Management Journal

2008; Herschcovis & Barling, 2010) and because wecontend that their effects begin very early in thesocialization experience.

Theory and research in other contexts suggestthat social support may have an important influ-ence on individuals’ hedonic tone, for the good (ifsupport is high) or bad (if low). The idea that hu-man beings will have especially strong emotionalreactions to supportive interactions with one an-other has been frequently noted in theories relatedto mood formation (e.g., Elfenbein, 2008; Watson,2000). Indeed, some have proposed that, as socialanimals, humans have a basic need for supportiveinteractions (Baumeister, 2005). Supporting thiscontention, studies have demonstrated the powerof supportive interactions on moods in daily life(Rafaeli, Cranford, Green, Shrout, & Bolger, 2008;Vinokur & Van Ryn, 1993; Watson, Clark, McIntyre,& Hamaker, 1992). First impressions exert a pow-erful influence on an individual’s beliefs and atti-tudes about a workplace (Stohl, 1986), so we expectthat initial levels of supervisor and coworker sup-port will be found to have a relationship to earlylevels of newcomer hedonic tone.

Hypothesis 1. Initial levels of supervisor andcoworker support are positively related to ini-tial levels of newcomer hedonic tone.

In addition to hypothesizing a relationship be-tween initial levels of supervisor and coworkersupport and newcomer hedonic tone, we also ex-amine what happens when support of a newcomerdecreases (or increases) over time. Coworkers andsupervisors may be more inclined to provide sup-port to newcomers early on, since it is generallyexpected that coworkers and supervisors will greeta newcomer and provide them with necessary in-formation. As time passes, however, we expect thatsome coworkers and supervisors will sustain sup-port, whereas others will not, even within the first90 days of a newcomer’s organizational entry. Forwhatever reasons (e.g., assumptions that the newemployee has settled in; lack of time), early offersof help and inclusion may dwindle in some workenvironments, while continuing in others. New-comers who are welcomed with high levels of sup-port may experience changes in hedonic tone ifsupport is progressively withdrawn, even thoughthey may have expected (or needed) it to continue.In this way, withdrawal of support can also beclassified in AET as an event (i.e., as a change fromone circumstance to another) with emotional con-sequences. Newcomers who perceive decreases in

support over time may interpret this decrease per-sonally, thinking they did something wrong, orthey may feel unnoticed, unimportant, and withoutan outlet for questions.

In other situations, individuals may start theirjobs with low perceptions of support. Researchershave noted stories of individuals beginning theirfirst day at work receiving little fanfare as to itsbeing stressful or negative experience (e.g.,Wanous, 1992). In some work environments, sup-port will remain low, whereas in others it may growas newcomers begin to meet individuals in theirwork environment. In either circumstance, our the-oretical framework suggests that the change in ex-perience catalyzes the process of event significanceappraisal and the subsequent evaluation of howpositive/negative that event was for a newcomer(Elfenbein, 2008; Weiss & Cropanzano, 1996). Spe-cifically, we expect that individuals who feel thatsupport is decreasing over time will have lowerlevels of hedonic tone over time, while those whofeel support is increasing over time will have im-proved levels of hedonic tone over time.

Hypothesis 2. There is a positive relationshipbetween trends in supervisor and coworkersupport and trends in newcomer hedonic toneover a newcomer’s first 90 days ofemployment.

Social undermining is anger or dislike directed atan individual, criticism of the individual’s charac-teristics, actions, or behavior, or actions that hinderthe individual’s goal attainment (Vinokur & VanRyn, 1993). Social undermining is not the same asan absence of support, because it involves an activeset of behaviors that make the recipient’s situationworse than it would be otherwise (Duffy, Ganster, &Pagon, 2002). In this sense, a person who abstainsfrom support could be conceived of as neglectfulrather than undermining, whereas a person whoabstains from undermining could be conceived ofas innocuous rather than supportive. From this per-spective, it is possible for an individual to demon-strate both high support and high undermining(e.g., be extremely helpful in some regards but en-gaging in negative behaviors in others); high sup-port and low undermining; low support and highundermining; or low both support and undermin-ing (e.g., neither helpful nor harmful, but simply“there”).

Although many studies have supported the ideathat undermining and related negative behaviorstoward newcomers and subordinates exist in work-


places (e.g., Chiaburu & Harrison, 2008; Duffy et al.,2002; Herschcovis & Barling, 2010), we are un-aware of any investigations of undermining in thecontext of newcomer adjustment. Yet such researchis important, as Ashforth, Sluss, and Saks indicatedwhen they called for research to examine psycho-social dynamics that can undermine the adjust-ment of newcomers (2007: 460). Undermining maybe directed at a newcomer for myriad reasons, in-cluding jealousy, lack of trust, fear that the new-comer’s actions will result in unwanted changes atwork, or simply a bad start. We expect that socialundermining events experienced by newcomerswill be related to lower levels of hedonic tone.Since newcomers are in a precarious state of tryingto feel welcome and get up to speed, such negativebehaviors will obstruct goal achievement (see, e.g.,Vinokur, Price, & Caplan, 1996) and will frustrate,irritate, and discourage the newcomers. In terms ofAET, newcomers attend to undermining from oth-ers as salient events that can impact their perfor-mance, and they will therefore initially react withnegative feelings when they see that such behavioris meant to hinder their progress. As with support,we expect that first impressions of underminingwill be related to more negative hedonic tone fromthe start of the employment experience.

Hypothesis 3. Initial levels (intercepts) of su-pervisor and coworker undermining are nega-tively related to initial levels of newcomerhedonic tone.

Similar to the proposed relationships betweentrends in support and hedonic tone, a relationshipbetween trends in undermining and hedonic toneis also expected. In some work environments, un-dermining may increase over time as colleaguesinteract more frequently with newcomers. As timepasses, initial patience and attempts to portray arosy picture to them may subside (Schein, 1978).Collaborations and interactions increase, allowingconflicting opinions and competition for territoryand recognition to surface. Individuals who per-ceive increases in undermining behavior may cometo feel that members of their social environment arerejecting and bullying them, which can lead toincreasingly negative moods. As Duffy et al. noted,“social undermining behaviors are insidious, inthat they weaken gradually or by degrees” (2002:332). In this sense, it is not only a person’s firstimpressions of social undermining that matter, butalso a continual trajectory of either constant orincreasing undermining from supervisors and co-

workers that impacts moods negatively. Theoryand research support the contention that negativeevents generate more salient and lasting emotionalimpact than do positive events (Elfenbein, 2008;Miner, Glomb, & Hulin, 2005; Weiss & Beal, 2005).Therefore, it is important to understand not onlythat negative events are related to negative mood,but also how such negative interactions factor intolong-term adjustment.

On the other hand, some newcomers may expe-rience a decrease in undermining, as an initial pe-riod of initiation or wariness subsides (Reichers,1987). Or it may simply take time for some cowork-ers and supervisors to come to trust and respect anewcomer as a fellow employee. Our theoreticalframework suggests that this change in circum-stances from higher to lower levels of criticism islikely to be evaluated positively, as the removal ofsuch aversive stimuli is negative reinforcement thatwill help facilitate goal success (Stein, Trabasso, &Liwag, 1993). To reiterate, the important consider-ation here is change over time as a result of recur-ring social interactions (i.e. events): AET posits,and research corroborates the assertion that the ab-sence of unpleasant events does not necessarilygenerate positive emotional reactions (Stallings,Dunham, Gatz, Baker, & Bengtson, 1997). Our ex-pectation is that declines in undermining will beassociated with recovery in positive emotionsover time.

Hypothesis 4. There is a negative relationshipbetween trends in supervisor and coworker un-dermining and trends in newcomer hedonictone over a newcomer’s first 90 days ofemployment.

Support, Undermining, and ProactiveSocialization Behavior

Newcomer reactions to a work environment arepartially the result of environmental social influ-ence but can also be the result of a newcomer’s ownproactive socialization behavior (e.g., Ashford &Black, 1996; Kammeyer-Mueller & Wanberg, 2003).Organizations and coworkers are incapable of pro-viding newcomers with all the information theyneed in a new work environment, and therefore it isimportant for newcomers to enact certain effortsthat will facilitate their own adjustment (Miller &Jablin, 1991). However, proactive socialization be-havior is also socially costly, bringing the potentialfor rejection or embarrassment if established organ-


ization members do not support a newcomer (Ash-ford, Blatt, & Vande Walle, 2003; Bolino, Valcea, &Harvey, 2010; Grant & Ashford, 2008). These pro-active efforts include establishing social relation-ships with others and seeking feedback regardingwork performance.

Theorists typically posit that proactivity is aproduct of a newcomer’s own initiative, but new-comers who receive higher levels of support feelmore comfortable with their work environment andwill therefore exert greater efforts toward proac-tively fitting in, much as those who experiencehigher levels of organizational socialization tacticsengage in higher levels of proactive socialization(Gruman, Saks, & Zweig, 2006). Insiders, who aremore comfortable with the social milieu, are in anexcellent position to take the first steps to approacha newcomer and invite him/her in. Such an ini-tially supportive attitude may foster feelings of con-fidence in the newcomer, which will encouragehim/her to reciprocate this positive social interac-tion in the form of increased proactive socializa-tion. When existing organization members providesupport, it creates an environment of psychologicalsafety (Edmondson, 1999; Schein & Bennis, 1965)that promotes newcomer proactive socialization. Inother words, a safety net of frequent socially sup-portive interactions enables people to make them-selves subsequently vulnerable to others (McAllis-ter, 1995). Individuals need such an effort fromsupervisors and coworkers to be able to admit gapsin knowledge, or to know that it is okay to askquestions and develop relationships. Those whofeel high initial support from coworkers and super-visors are likely to feel that their work group wantsthem to succeed, which will in turn lower the per-ceived costs often associated with proactive behav-ior (e.g., Parker, Williams, & Turner, 2006).

Hypothesis 5. Initial levels of supervisor andcoworker support are positively related to ini-tial levels of newcomer proactive socializationbehaviors.

Miller and Jablin noted that “newcomers havereported to us that they become increasingly hesi-tant to ask questions of co-workers because theyfeel like they are ‘bugging’ them and fear beingreprimanded or ‘cut off’ from future information”(1991: 97). Continuation of social support can senda signal that a newcomer is not being overly solic-itous of help and assistance and in this way canencourage more proactivity over time. When new-comers experience higher levels of social support

over time, they will feel increasingly validated andwill expect that their efforts to fit into the workenvironment will be rewarded. In particular, asnewcomers become more integrated into the socialenvironment, expectations for their performancewill become more exacting (Fisher, 1986), so main-tenance of social support during this crucial periodis extremely important for encouraging continuedproactive socialization behaviors. On the contrary,we anticipate that newcomers who believe theirsupervisors and coworkers are withdrawing socialsupport over time will respond in kind by alsoreducing their initiative to fit into the workplace.

Hypothesis 6. There will be a positive relation-ship between trends in supervisor and coworkersupport and trends in newcomer proactive so-cialization behavior over a newcomer’s first90 days of employment.

Undermining a newcomer, especially in the ini-tial weeks of work, will send a strong signal that heor she is not welcome (Baron, 1988). Social under-mining also provides negative feedback to the new-comer about his or her fit with a work group and anorganization, which will reduce the newcomer’sself-efficacy (Duffy et al., 2002) and will thereforedampen the newcomer’s motivation to proactivelyengage her/his environment. Under these condi-tions, it is likely that the newcomer will reciprocatethese overt efforts to exclude or undermine his orher position by withdrawing from the work groupand engaging in fewer proactive efforts to fit in.

Hypothesis 7. Initial levels of supervisor andcoworker undermining are negatively relatedto initial levels of newcomer proactive social-ization behaviors.

Initial levels of social undermining are likely tonegatively influence levels of newcomer proactiv-ity, but changes in social undermining will also beassociated with changes in newcomer proactivity.A newcomer who experiences a positive trend inundermining is likely to feel that the environmentis becoming progressively less and less welcomingand therefore will have a corresponding decreasingtrend in proactive socialization. Following a pat-tern much like what we propose for hedonic tone,those who feel that undermining is constant orincreasing will have more negative reactions interms of proactive socialization behaviors thanthose who see undermining decrease over time.


Hypothesis 8. There is a negative relationshipbetween trends in supervisor and coworker un-dermining and trends in newcomer proactivesocialization behaviors over a newcomer’s first90 days of employment.

Relationships with Socialization Outcomes

We posit that these processes have importantconsequences for newcomer adjustment to workenvironments. To assess newcomer adjustment, wefocus on work proactivity, social integration, organ-izational commitment, and withdrawal behaviorsbetween 90 and 100 days after organizational entry,as well as actual turnover (exit) one year after ourfinal survey period. Our choice of socialization out-comes matches that of prior research, which hasidentified social integration and commitment(Bauer et al., 2007; Fisher, 1986), as well as workwithdrawal and remaining with an organization(Kammeyer-Mueller & Wanberg, 2003), as indica-tions of positive newcomer adjustment. Our inclu-sion of work proactivity as an outcome in a studythat incorporates proactive socialization respondsto a call for better integration of proactivity across avariety of domains (Grant & Ashford, 2008). In con-trast to proactive socialization behaviors (e.g., pro-activity related to socialization, such as developingrelationships and seeking feedback [Ashford &Black, 1996]), work proactivity relates to the extentto which employees make suggestions for im-proved efficiency in their work units and becomeinvolved with ways to increase the effectiveness oftheir organization (Griffin, Neal, & Parker, 2007).

Theoretically, we anticipate that the effects ofsupport and undermining on these work outcomeswill be mediated through hedonic tone. The role ofhedonic tone in newcomer socialization has yet tobe explored, but there are reasons to expect that itmay be a critical variable in the adjustment process.According to the broaden-and-build theory(Fredrickson, 2001), individuals in more positivemoods are motivated to expand resources and pur-sue goals, which may facilitate work. Affectiveevents theory and other related models of emo-tional processes in organizations propose that fol-lowing exposure to an emotion-eliciting event,changes in affect will ultimately result in changesin attitudes (Elfenbein, 2008; Weiss & Cropanzano,1996). In socialization models (e.g., Bauer et al.,2007; Fisher, 1986), feelings of social integration (atthe work group level) and organizational commit-ment (at the organizational level) are the most rel-

evant job attitudes to assess for newcomers. Wealso anticipate that hedonic tone may influencenewcomers’ desire to actively contribute to andremain in an organization. Individuals in more neg-ative affective states tend to have higher levels ofwithdrawal and turnover intentions in both cross-sectional research (e.g., Cropanzano, James, & Kon-ovsky, 1993) and daily diary studies (e.g., Grandey,Tam, & Brauburger, 2002), likely because these in-dividuals wish to remove themselves from situa-tions and environments in which these negativefeelings tend to occur. For newcomers, who havelittle stake in an organization in which they haveonly been working for a short period of time, thisdesire to leave may be especially salient, as theyarguably have less to lose than others who haveinvested more time and personal resources intotheir jobs.

Hypothesis 9. Hedonic tone acts as a mediatorbetween supervisor and coworker support andundermining and the socialization outcomesof work proactivity, social integration, organi-zational commitment, withdrawal behaviors,and voluntary turnover.

Besides the relationship between hedonic toneand adjustment outcomes, newcomer proactive so-cialization behaviors will also mediate betweensupport and undermining and important work out-comes. Socialization theory posits that newcomerswho engage in more proactive behaviors will havea better sense of their work environment and theirrole in it (Miller & Jablin, 1991). Those who proac-tively seek out feedback and build relationshipswill have both knowledge and social resources tobehave more proactively in their work roles thanthose who do not have this knowledge and socialresource base. Proactive newcomers will also feelthey are a part of their organization as a wholebecause they will understand more about their en-vironment (Ashford & Black, 1996; Morrison,1993), which should in turn increase their feelingsof organizational commitment. Efforts to proac-tively build relationships and seek out informationshould also alleviate feelings of social separation,confusion, or alienation, which can lead to workwithdrawal (Hanisch & Hulin, 1991).

Hypothesis 10. Proactive socialization behav-iors act as a mediator between supervisor andcoworker support and undermining, and thesocialization outcomes of work proactivity, so-


cial integration, organizational commitment,withdrawal behaviors, and voluntary turnover.

METHODS

Data and Sample

The initial pool of participants consisted of 906new hires at a major research university in theupper Midwest. Potential participants were identi-fied from organizational hiring records. Respon-dents received an e-mail and follow-up phone callinviting them to participate in the online survey.The average age of respondents was 34.27 years(s.d. � 9.54), and the average number of years ofprofessional experience (in any occupation) was8.93 (s.d. � 9.19). Sixty-five percent were female,46.4 percent were married, and 77.3 percent werewhite. Finally, 49.8 percent of respondents held agraduate or professional degree, indicating this wasa highly educated sample. Common occupationsfor participants included research specialist, tech-nician, academic advisor, coordinator, teachingspecialist, and accounting.

We did not include faculty in our sample. Ourfocus on support and undermining requires a de-gree of regular interaction with coworkers and su-pervisors. At this university, faculty are not re-quired to work in their offices, whereasadministrative employees work in an office withother employees. Telecommuting is quite commonamong faculty members but is less so for adminis-trative personnel (Ng, 2006). Furthermore, unlikethe general workforce, faculty members do not usu-ally interact with their supervisors on a regularbasis (Olsen, 1993), nor do they regularly work ingroup settings. Further, faculty differed from thegeneral workforce in the extent of formal training orinstitutionalized socialization they experience(Jones, 1986).

Of the 906 newcomers e-mailed to solicit partic-ipation, 316 completed the introductory survey(34.9% initial response rate). The first survey wasdistributed within individuals’ first two weeks ofemployment. Weekly surveys were distributed overthe next three months, for a total of 14 possiblesurveys. Our sample was limited to those whoworked over 30 hours per week, since part-timeworkers may differ in their socialization processes.This screen eliminated 25 respondents. An addi-tional 8 individuals dropped out of the study afterthe introductory survey. Thus, 264 participantswere included in the weekly ongoing survey pool.

Participants were tracked in cohorts based on thedate they were hired; in each week we obtained anupdated list from hiring records for the new cohortstarting work that week. Our sample consisted of 37weekly cohorts for which data were obtained overour total sampling period of 51 weeks. Of the 264respondents included in the ongoing pool, 255(28.1% overall response rate, 97% ongoing studyretention rate) were surveyed through all 14rounds. The 30-person discrepancy from the begin-ning to end of the survey periods was due to 30individuals being removed from the sample for fail-ing to complete at least 11 of the 14 surveys on time(within the survey week).

The use of the three-month time frame (i.e., sur-veys taking place over the first three months of thenewcomer experience) matched the often statedconviction that what matters in “onboarding” is thefirst 90–100 days (Bradt, Check, & Pedraza, 2006;Watkins, 2003). Research has furthermore shownthat the first three months of a newcomer’s tenurerepresent the greatest amount of change in workattitudes (Lance et al., 2000). Weekly rather thanless frequent assessments were used to allow us toadequately model the decline or increase in ourvariables over this three-month period. Finally, incontrast to less frequent assessments, one-week in-tervals are discrete and easy to recall.

Measures

All independent variables were collected via aweekly internet-based survey.

Independent variables. Our measures for socialsupport and undermining were taken from Vinokuret al. (1996). Social support was measured with fiveitems regarding supervisors and five items regard-ing coworkers that asked whether they had engagedin behaviors such as “provided you with encour-agement,” “listened to you when you needed totalk,” or “helped you understand and sort thingsout” over the past week. Undermining was mea-sured with five items regarding supervisors andfive items regarding coworkers that asked aboutwhether they had acted in ways such as “madeyour life difficult,” “acted in ways that show theydislike you,” or “criticize you” over the past week.Responses ranged from 1 (“to no extent”) to 5 (“to agreat extent”).

Hedonic tone was measured using a scale devel-oped by Scollon et al. (2005). This scale measuresaffect with eight adjectives (four for pleasant mood,four for unpleasant mood). Example adjectives in-


clude “happiness,” “liking/affection,” “irritation”(reverse-scored), and “sadness” (reverse-scored).Respondents indicated the extent to which theyhad felt this way over the last week at work (1,“very slightly or not at all,” to 5, “extremely”). Highscores indicate positive hedonic tone.

We also included the newcomers’ self-reportedproactive socialization behavior, using items pre-sented in Ashford and Black (1996). Four itemswere used to assess information/feedback seeking(e.g., “To what extent have you sought feedback onyour performance after assignments?”); three itemsassessed general socializing with coworkers (e.g.,“To what extent have you attempted to form socialrelationships with co-workers?”); and three itemsassessed relationship building with one’s boss (e.g.,“To what extent have you tried to spend as muchtime as you could with your boss?”). Evidence sup-porting the scales’ internal consistency and validityis provided in Ashford and Black (1996). Responseswere answered with respect to the past week onscales ranging from 1 (“to no extent”) to 5 (“to agreat extent”).

Dependent variables. With the exception ofturnover (which was measured one year after theend of the study), our work outcome variables weremeasured in the last survey round, on a separateoccasion from the measurement of the independentvariables and mediators. For all work outcome vari-ables, responses were on a scale ranging from 1,“strongly disagree,” to 5, “strongly agree.”

Work proactivity was measured with nine items(Griffin et al., 2007). Each began “How often in thelast week have you,” followed by, for example,“initiated better ways of doing your core tasks,”“improved the way your work unit does things,”and “come up with ways of increasing efficiencywithin the organization.”

Social integration was measured with sevenitems answered with respect to the last week (Chao,O’Leary-Kelly, Wolf, Klein, & Gardner, 1994; Mor-rison, 1993). Examples include “My coworkersseem to accept me as one of them,” and “Within mywork group, I would be easily identified as ‘one ofthe gang.’”

Organizational commitment was measured withthe nine-item version of the Organizational Com-mitment Questionnaire (Mowday, Steers, & Porter,1979). Example items include “I find that my val-ues and this organization’s values are very similar”and “I really care about the fate of this organiza-tion.” Withdrawal behavior was measured with aten-item scale assessing the frequency of engaging

in a variety of withdrawal behaviors (Roznowski &Hanisch, 1990). Example items include “I often failto attend scheduled meetings,” “I often let othersdo my work for me,” and “I am often absent fromwork.” Voluntary turnover was assessed one yearafter the last survey through archival organizationalrecords of whether the employee had voluntarilyterminated his or her employment and the date ofthe last day of work.

Control variable. Because research suggests dis-positions affect individuals’ perceptions of nega-tive behaviors from others (Bowling & Beehr, 2006),proactivity (Grant & Ashford, 2008), and job atti-tudes (Thoresen, Kaplan, Barsky, Warren, & deChermont, 2003), we included a measure of trait(stable) neuroticism collected in the first week ofemployment in all analyses. The source of thisscale was the International Personality Item Pool(Goldberg, 1999). Respondents indicated the extentto which statements such as “worry about things”and “have frequent mood swings” described them.Responses ranged from 1 (“strongly disagree”) to 5(“strongly agree”).

Analyses

All analyses except those for turnover were esti-mated using latent growth modeling (Chan &Schmitt, 2000). Because the number of surveyrounds made our model extremely complex, we fitmeasurement models independently of our struc-tural model. We used tests of measurement invari-ance by comparing models with factor loadings andintercepts of items or item parcels (for scales withmore than five items) in terms of fixed versus freeas described by Chan (1998). Comparing CFI andRMSEA values between the free and constrainedmodels at two decimal places yielded no differ-ences, suggesting that measurement invariance is atenable assumption. We investigated the form ofgrowth trajectories of our time-varying predictors(supervisor and coworker support and undermin-ing) by estimating both linear and nonlinear mod-els (e.g., Jokisaari & Nurmi, 2009). All models wereestimated with Mplus 6.1 (Muthén & Muthén,2010). Model fit was assessed with chi-square, CFI,RMSEA, and SRMR.

Initially, we fit linear models by setting the coef-ficients of all variables to 1 on the intercept factors(time 1 survey variables) and to 0 through 11 on theslope factors (survey variables over time). Thesemodels showed poor fit when estimated for eachvariable individually and failed to converge when


estimated as a unified structural model. Thus, weestimated a nonlinear model that allowed the pa-rameters to freely vary within a range from 0 (firstround) to 1 (last round). These nonlinear specifica-tions fit the data comparatively well (�2 � 4,772, df� 2,642; RMSEA � .05, CFI � .90, and SRMR �0.07). Examination of the slope coefficients andassociated predicted values for each variable overtime demonstrated that supervisor support, co-worker support, and hedonic tone all followed non-linear trajectories that decreased more substantiallyat first, and then leveled off somewhat. Overalltrends among participants in supervisor and co-worker undermining and proactivity were not sig-nificant, suggesting no consistent increases or de-creases in these variables over time among thesample as a whole. All slopes had significant vari-ance coefficients, demonstrating variability inslopes across individuals.

We analyzed turnover using hazard modeling(Morita, Lee, & Mowday, 1993) with repeated mea-sures of the independent variables (i.e., time-vary-ing covariates). In these analyses, job duration mea-sured in days is the dependent variable, and eachsurvey round is included as a predictor of turnover.Data on all respondents who did not voluntarilyquit prior to the final survey round were treated asright-censored.

RESULTS

Descriptive statistics, reliability information, andzero-order correlations among study variables are

presented in Table 1. The average of between-per-sons contemporaneous correlations across wavesare presented below the diagonal, with the relevantwithin-person correlations presented above thediagonal.

As shown in Table 2 and illustrated in Figure 2,the mean growth rate trajectories in supervisor sup-port, coworker support, and hedonic tone were sig-nificant and negative, suggesting that over the first14 weeks, supervisor and coworker support andhedonic tone declined. The mean growth rate func-tions for supervisor and coworker underminingwere not significant, suggesting no detectable aver-age change in these variables over time (e.g., seeFigure 2). For all variables, however, between-per-sons variance in initial status and growth rates wassignificant, so these variables can act as predictorsof between-persons differences in outcomes. Thecovariance between initial status and growth wassignificant for both supervisor support and super-visor undermining. Higher initial levels of supervi-sor support were associated with greater declinesin support over time. Higher initial levels of super-visor undermining were associated with greater in-creases in supervisor undermining over time.

Table 3 portrays the relationships between super-visor and coworker support and newcomer hedonictone. Hypothesis 1, suggesting initial levels of su-pervisor and coworker support are related to initiallevels of newcomer hedonic tone, was supported.After controlling for neuroticism, we found thatinitial levels of supervisory and coworker support,respectively, were positively related to initial lev-

TABLE 1Means, Standard Deviations, Reliabilities, and Correlationsa

Variables Mean

StandardDeviationBetween

StandardDeviation

Within 1 2 3 4 5 6 7 8 9 10 11

1. Neuroticism 2.49 .66 .912. Supervisor support 3.11 .76 .57 �.23 .94 �.07 .46 �.03 .24 .403. Supervisor undermining 1.16 .43 .24 .28 �.12 .93 �.02 .35 �.21 .084. Coworker support 3.39 .69 .48 �.26 .55 �.07 .92 �.02 .25 .335. Coworker undermining 1.18 .42 .23 .19 �.09 .58 �.14 .91 �.16 .016. Hedonic tone 3.71 .61 .33 �.45 .37 �.32 .42 �.31 .86 .187. Proactive socialization 2.64 .65 .42 �.17 .57 �.01 .52 .01 .36 .898. Work proactivity 3.38 .80 .03 .19 .05 .17 .08 .16 .32 .969. Social integration 3.93 .58 �.29 .23 �.11 .33 �.24 .38 .29 .09 .87

10. Organizational commitment 3.78 .66 �.30 .24 �.12 .30 �.12 .42 .28 .14 .29 .9211. Withdrawal behaviors 1.48 .51 .09 �.12 .17 �.24 .27 �.24 �.17 �.18 �.23 �.33 .86

a Coefficients below the diagonal represent the average between-person correlations across each survey wave for repeated measures;coefficients above the diagonal represent the average within-person correlation across survey waves. Average internal consistencyreliability estimates (alpha coefficients) are on the diagonal in italics.


TABLE 2Parameter Estimates of Latent Growth Modelsa

VariablesSupervisor

SupportSupervisor

UnderminingCoworkerSupport

CoworkerUndermining

HedonicTone

ProactiveAdjustment

Mean initial status 3.45** 1.13** 3.72** 1.17** 3.8** 2.64**Variance in initial status 0.58** 0.14** 0.35** 0.09** 0.32** 0.72**Mean growth rate �0.46** 0.03 �0.47** 0.01 �0.14** �0.05Variance in growth rate 0.38** 0.06** 0.34** 0.03** 0.11** 0.53**Covariance (initial status and growth rate) �0.17** 0.02* �0.05 0.01 �0.01 �0.15

a n � 264.* p � .05

** p � .01

FIGURE 2Support, Undermining, Affect, and Proactivity over Time


els of newcomer hedonic tone (� � .15, p � .05; � �.29, p � .01).

Hypothesis 2 predicts positive relationships be-tween the trends (i.e., slopes over time) of supervi-sor and coworker support and hedonic tone. Par-tially supporting this hypothesis, this relationshipwas statistically significant for supervisor support(� � .51, p � .05) but not coworker support. Thesignificant finding shows that within persons, asindividuals experienced increases in supervisorysupport over time, they also reported more positivelevels of hedonic tone. Similarly, as individualsexperienced decreases in supervisory support overtime, they reported more negative levels of hedo-nic tone.

Hypothesis 3, suggesting initial levels of super-visor and coworker undermining are related to ini-tial levels of newcomer hedonic tone, was sup-ported only for coworker undermining. Initiallevels of coworker undermining were negativelyrelated to initial levels of newcomer hedonic tone(� � �.34, p � .01). Hypothesis 4, predicting neg-ative relationships between the trends (i.e., slopesover time) of supervisor and coworker support andhedonic tone, was not supported. Neither slope ofundermining was related to the slope in hedo-nic tone.

Table 4 portrays the relationships between super-visor and coworker support and newcomer proac-tive socialization behaviors. Supporting Hypothe-sis 5, we found a positive relationship between theinitial status of proactive socialization behaviorand both supervisor support (� � .50, p � .01) and

coworker support (� � .28, p � .01). SupportingHypothesis 6, we also found positive relationshipsbetween the trends of supervisor and coworker sup-port and proactive socialization behavior over time (�� .42, p � .01; � � .23, p � .01). This finding suggeststhat as individuals experienced increases (or de-creases) in supervisory or coworker support overtime, they also reported increased (or decreased) lev-els of proactive socialization behavior.

Hypothesis 7 posits that higher initial levels ofundermining are related to lower initial levels ofproactivity, while Hypothesis 8 suggests that morepositive trends in undermining would predict de-creases in proactive socialization. However, resultsindicated that the neither the intercepts nor theslopes of undermining were significantly predic-tive of proactive socialization, and so neither Hy-pothesis 7 nor Hypothesis 8 was supported.

Next, we examined Hypotheses 9 and 10—thusinvestigating whether hedonic tone and proactivebehaviors mediate the relationship between sup-port and undermining and socialization outcomes(work proactivity, social integration, organizationalcommitment, withdrawal behaviors, and voluntaryturnover). Direct effects (support, undermining,and the mediators predicting socialization out-comes) are shown in Table 5. Indirect effects (sup-port and undermining mediated by hedonic toneand proactivity) are shown in Table 6. We note thatwhile the indirect effects were tested with the di-rect effects also included in Figure 1, the arrows fordirect effects are not presented, to make the imageeasier to read. Finally, results for voluntary turn-over are reported separately, in Table 7, since this

TABLE 3Predictors of Hedonic Tonea

Variables

InitialStatus

(Intercept)of Hedonic

Tone

Change(Slope)

in HedonicTone

Neuroticism �0.54** 0.12Supervisor support intercept 0.15*Supervisor support slope 0.51**Supervisor undermining intercept 0.07Supervisor undermining slope �0.06Coworker support intercept 0.29**Coworker support slope 0.08Coworker undermining intercept �0.34**Coworker undermining slope �0.13

a n � 264.* p � .05

** p � .01

TABLE 4Predictors of Proactive Socializationa

Variables

InitialStatus

(Intercept)of ProactiveSocialization

Change(Slope)

in ProactiveSocialization

Neuroticism �0.06 0.04Supervisor support intercept 0.50**Supervisor support slope 0.42**Supervisor undermining intercept 0.09Supervisor undermining slope �0.04Coworker support intercept 0.28**Coworker support slope 0.23**Coworker undermining intercept 0.05Coworker undermining slope 0.02

a n � 264.** p � .01


variable required a different analytical approach(i.e., hazard modeling).

Table 5 portrays several significant direct effects.Particularly of interest is the differential pattern ofrelationships found for coworker support and un-dermining relative to supervisor support and un-dermining. For instance, initial supervisor under-mining (at the start of employment) was positivelyrelated to social integration 90 days later (� � .29, p� .01), yet the intercept for initial coworker under-

mining was negatively related to social integrationat 90 days (� � �.38, p � .01). Moreover, we foundthat while initial levels of supervisor support werepositively related to withdrawal behaviors such asabsenteeism (� � .15, p � .05), a negative pathcoefficient was found for coworker support (� ��.18, p � .05). Finally, results suggested that initiallevels of hedonic tone and both initial levels andthe slope of socialization proactivity were predic-tive of work proactivity (see Table 5).

TABLE 5Direct Effects Predicting Socialization Outcomesa

VariablesWork

ProactivitySocial

IntegrationOrganizationalCommitment

WithdrawalBehaviors

Neuroticism 0.27** �0.13 0.04 �0.11Supervisor support intercept �0.05 �0.08 �0.09 0.15*Supervisor support slope 0.08 �0.01 0.10 �0.10Supervisor undermining intercept �0.02 0.29** 0.05 �0.10Supervisor undermining slope �0.12 0.04 �0.09 0.00Coworker support intercept �0.03 0.10 0.11 �0.18*Coworker support slope 0.02 0.07 �0.05 0.02Coworker undermining intercept 0.22 �0.38** 0.04 0.27**Coworker undermining slope �0.09 �0.11 �0.11 0.09Hedonic tone intercept 0.19* 0.10 0.32** �0.07Hedonic tone slope �0.07 0.05 �0.02 0.04Proactive socialization intercept 0.33** 0.11 0.15* �0.13*Proactive socialization slope 0.17* 0.05 �0.02 0.04

a n � 264.* p � .05

** p � .01

TABLE 6Indirect Effects of Support and Undermining Predicting Socialization Outcomesa

VariablesWork

ProactivitySocial

IntegrationOrganizationalCommitment

WorkWithdrawal

Supervisor support intercept Hedonic tone intercept 0.03 0.02 0.05* �0.01Proactivity intercept 0.16** 0.06 0.07* �0.07*

Supervisor support slope Hedonic tone slope �0.04 0.03 0.00 0.02Proactivity slope 0.07 0.02 0.00 0.02

Supervisor undermining intercept Hedonic tone intercept 0.01 0.01 0.02 �0.01Proactivity intercept 0.03 0.01 0.01 �0.01

Supervisor undermining slope Hedonic tone slope 0.00 0.00 0.00 0.00Proactivity slope �0.01 0.00 0.00 �0.01

Coworker support intercept Hedonic tone intercept 0.05 0.03 0.10** �0.02Proactivity intercept 0.09** 0.03 0.04* �0.04**

Coworker support slope Hedonic tone slope �0.01 0.00 0.00 0.00Proactivity slope 0.04 0.01 �0.01 0.01

Coworker undermining intercept Hedonic tone intercept �0.06 �0.04 �0.11 0.02Proactivity intercept 0.02 0.01 0.01 �0.01

Coworker undermining slope Hedonic tone slope 0.01 0.00 0.00 �0.01Proactivity slope 0.00 0.00 0.00 0.00

a n � 264.* p � .05

** p � .01


Results in Tables 6 and 7 show partial support forHypotheses 9 and 10. As shown in Table 6, initialsupervisor and coworker support were positivelyrelated to all four socialization outcomes throughthe two mediators (hedonic tone and proactive so-cialization behavior). The relationship between ini-tial levels of coworker undermining and reducedorganizational commitment 90 days later was alsomediated by hedonic tone and proactive socializa-tion behavior.

Table 7, reporting the repeated-measures turn-over analyses described earlier, show that supervi-sor undermining over time was a significant pre-dictor of the speed and occurrence of newcomerturnover. Those employees who reported greaterlevels of undermining from their supervisors vol-untarily left their positions sooner than those whodid not report experiencing such behaviors (hazardratio � 3.01, p � .01). Interestingly, the turnoverhazard was not statistically significant for reportedundermining from coworkers (hazard ratio � 0.53,n.s.). Neither proactive socialization nor hedonictone was related to this outcome. As such, therewas no evidence for the mediation hypothesis in-volving the outcome of turnover. Drawing on asuggestion from an anonymous reviewer, we alsoestimated models that used only the initial levels ofsupport and undermining as predictors of turnover,in keeping with the procedure described in Kam-meyer-Mueller, Wanberg, Glomb, and Ahlburg(2005). This model was not significant overall, norwere any of the individual coefficients, suggestingthat it is the pattern of undermining over time thatis associated with turnover rather than just the ex-perience of undermining in a first week.

DISCUSSION

Newcomer socialization involves repeated inter-actions of newcomers with coworkers and supervi-sors. Yet a substantial amount remains to belearned about the power and meaning of initial anddynamic social interactions for organizational new-comers. Our study provides a unique glimpse intothese processes, exploring how perceptions of su-pervisor and coworker support and underminingare important for newcomer adjustment both ini-tially and over time. In the present investigation,we aimed to assess how social support and under-mining are related to newcomer hedonic tone andproactive socialization behaviors and how hedonictone and proactive socialization, in turn, act asmediators between support and undermining and avariety of socialization outcomes. In examiningthese hypotheses, we offer three major contribu-tions. First, our incorporating both supervisor andcoworker support into common models to test theirindependent and complementary effects corre-sponds with a growing body of research that sug-gests that both supervisors and coworkers are crit-ical to the social environment for employees (e.g.,Chiaburu & Harrison, 2008; Duffy et al., 2002;Herschcovis & Barling, 2010). Second, we exam-ined whether early entry undermining and sup-port, as well as changes over time in these con-structs, are related to adjustment outcomes fororganizational newcomers. Third, we exploredthe mechanisms by which support and under-mining might be related to a variety of indicatorsof newcomer adjustment.

Implications for Theory and Research

As we noted in the introduction, two key areas ofcontroversy motivated this investigation. The firstissue involves interpersonal interactions as a po-tential source of newcomer adjustment patterns.Our results clearly show that contributors to thesocialization literature should not overlook the roleof interpersonal interactions when conceptualizinghow newcomers become adjusted to their newwork environments. Theories of newcomer social-ization will need to incorporate both support andundermining and might well incorporate ideasfrom the literature on interpersonal relationshipsthat explores some of these same issues (e.g., Vi-nokur & Van Ryn, 1993). Moreover, although somemodels position newcomer proactivity as largelythe result of newcomers’ own initiative (e.g., Miller

TABLE 7Prediction of Voluntary Turnover Hazarda

Variables

Voluntary Turnover

HazardRatio z

HazardRatio z

Neuroticism 1.58 1.41 1.38 0.95Supervisor support 0.87 �0.48 0.81 �0.65Supervisor undermining 3.01 3.22** 2.79 2.96**Coworker support 1.10 0.28 1.08 0.23Coworker undermining 0.53 �1.37 0.51 �1.47Proactive socialization 1.43 0.97Hedonic tone 0.66 �1.09Wald �2 (df) 14 (5) 16 (7)

a n � 264.** p � .01


& Jablin, 1991), here it appears that proactivity ismore likely when the appropriate initial social en-vironment is present together with a pattern of con-tinued support. This model is similar to the inter-actionist model of employee proactivity putforward by a number of scholars (e.g., Grant & Ash-ford, 2008; Parker et al., 2006) that positions pro-active behavior as the result of both of personal andsituational determinants. Prior research in a sampleof experienced employees has demonstrated thatabusive supervision was related to turnover (Tep-per, 2000). We extend this research with resultsshowing that these relationships occur quite earlyin the organizational entry period and are onlysignificant for undermining by supervisors.

The second issue involved the dynamic relation-ships among socialization variables. Our resultsportray clearly that what happens initially—firstimpressions—is very important to newcomers.There is a substantial literature on the importanceof first impressions as a catalyst for forming atti-tudes toward other individuals through a process ofconfirmatory bias (e.g., Dougherty, Turban, & Cal-lender, 1994; Tetlock, 1983). Our study suggeststhat first impressions may also serve to develop anorganizational newcomer’s impression of the socialenvironment in her/his new workplace as a whole,as Stohl (1986) suggested. Future research mightaddress this theoretical area by examining whethernewcomers seek out information that is consistentwith their first impressions of support and/orundermining.

Our investigation provides further informationon the extent to which changes in support andundermining are important over time. Our resultssuggest that newcomers continue to need supportover time as they become acclimated (e.g., Fisher,1986; Wanous, 1992). The effects of support overtime suggest that colleagues can, at least in part,make up for a “slow start.” At the same time, thepositive relationship between the slope of under-mining and withdrawal behaviors at the end of thestudy suggests that some work groups start posi-tively, and undermining increases little by littleover time. Such increases are related to higherwithdrawal behaviors on the part of the newcomer.Most theory on newcomer adjustment has not spe-cifically taken a position on how changes over timein a social environment might influence newcom-ers, so further work following the direction laid outby Ashforth (2012) to deal with socialization dy-namics will be informative.

It is worth noting that our hypotheses regardingthe negative influence of undermining on proactivesocialization were not supported. One anonymousreviewer suggested that some individuals who areundermined may engage in more proactive social-ization in an effort to overcome the active efforts toblock their goal progress. Our results could thuspotentially indicate that countervailing forces acton undermined newcomers and that some under-mined newcomers engage in more socialization towork against the fact that they’ve been under-mined, whereas other newcomers engage in lesssocialization because they feel rejected anddispirited.

Finally, our conceptual model addressed themechanisms (hedonic tone and proactive socializa-tion behavior) by which newcomer support andundermining are posited to relate to important so-cialization outcomes, including organizationalcommitment, work proactivity, social integration,work withdrawal, and turnover. Although AET wasused as a theoretical framework to explain the re-lationship between support and undermining andthe terminal work outcomes of proactivity, commit-ment, social acceptance, and withdrawal, our datatend to show that hedonic tone intercepts or slopesdo not act as mediators but do have some maineffects. Instead, our results clearly suggest that theintercept of proactivity was the most important me-diating variable, affecting work proactivity, organ-izational commitment, and work withdrawal. Thispattern is consistent with theories of proactive so-cialization that propose that newcomers who takean active role in their own adjustment tend to havesuperior outcomes (e.g., Miller & Jablin, 1991).

Limitations and Future Research Directions

Our study also adds to the organizational social-ization literature by demonstrating the combinedeffects of coworker and supervisor behaviors on avariety of outcomes. Prior research on socializationhas mostly focused on the influence of supervisors(e.g., Bauer, Erdogan, Liden, & Wayne, 2006; Bauer& Green, 1998; Jokisaari & Nurmi, 2009). Our studyalso showed that supervisors are important, butcoworker support and undermining exert addi-tional influence on outcomes. This pattern of find-ings matches those of prior research showing thatsupervisor and coworker social behaviors are re-lated to numerous work outcomes (Chiaburu & Har-rison, 2008; Herschovis & Barling, 2010). Futureresearch might build on this foundation by exam-


ining the outcomes of different aspects of support,such as the differences between affective and in-strumental support found in previous research (Co-hen & Wills, 1985). For instance, affective supportmay be more strongly related to social integrationoutcomes than to core task outcomes, as insidersexhibit behaviors aimed at developing friendshipsand making newcomers feel as if they are nowinsiders themselves. In comparison, instrumentalsupport may be more associated with outcomes suchastask efficiency and work group performance.

Our results demonstrating the decline of super-visory and coworker support, as well as newcomerhedonic tone (within the first 90 days of employ-ment) as generalized trends are intriguing. Ourfinding extends findings by Boswell and colleaguesthat individuals tend to experience a decrease injob satisfaction over their first year of employment(Boswell, Boudreau, & Tichy, 2005; Boswell,Shipp, Payne, & Culbertson, 2009). In addition tofocusing on a mood-based construct (hedonic tone),our findings indicate that declines in hedonic toneare evident within a much more defined time frameand with the use of weekly assessments. Althoughthe employees in our study belonged to a diversearray of work groups and environments, futurestudies will be needed to examine the generaliz-ability of our findings. Researchers might endeavorto understand why declines in support occur andexamine further to what extent these declines insupport are a concern. It seems from our results thatit would be valuable for organizations to preventsuch declines in support; other consequences ofdeclining support would be valuable to identify.

Newcomers may be particularly vulnerable asthey try to make sense of an unfamiliar work envi-ronment (Hogg & Terry, 2000; Wanous, 1992). Tothis end, future research should explore moderat-ing effects of support and undermining, respec-tively, on the relationship between early entrystressors and socialization outcomes. Specifically,how might supportive coworkers and/or supervi-sors help newcomers as they navigate their newwork roles, acquire new skills, and handle unper-formed demands? Having been in the same positionat one time themselves, coworkers may be espe-cially able to empathize with and support newcom-ers. Of equal interest, how might undermining inthe first weeks of a job interact with work stressorsto predict distinct patterns of socialization out-comes? A criticizing supervisor could make theexperience of typical stressors be instead felt asmuch more acute.

Future research might also look into individualdifferences that moderate the relationship betweenundermining and moods or proactivity. Althoughundermining doesn’t influence hedonic tone orproactivity in general, some employees may be es-pecially prone to experience sadness and persever-ate about rejection, and also to withdraw efforts tofit in. Social psychology researchers have identi-fied rejection sensitivity as an individual differencethat makes some people especially prone to havenegative reactions when rejected (Ayduk, Men-doza-Denton, Mischel, Downey, Peake, & Rodri-guez, 2000).

One limitation of our investigation is the use ofself-report scales for constructs, raising the possi-bility that common method variance is accountingfor our observed relationships (Podsakoff, MacKen-zie, Lee, & Podsakoff, 2003). However, a number offeatures of our investigation help to guard againstthis explanation. Specifically, we included con-trols for dispositional neuroticism, which mightinfluence perceptions of a variety of work features.Second, we separated our predictors and outcomesover time and estimated how changes in our dy-namic variables were related to one another. Third,we used an objective, behavioral turnover indicatoras a dependent variable in addition to our self-reported relationships. Future research could ex-amine coworker and supervisor perceptions of sup-port and undermining over time to see how theycorrespond with results from our self-report sur-vey. However, our focus on the newcomers’ percep-tions of these behaviors is consistent with findingsfrom Bennett and Robinson (2003), who urged re-searchers to examine deviance from a victim’sperspective.

Another limitation of our study was the lack of ameasure of pre-entry knowledge. Prior research hasindicated that individuals who report higherknowledge of an organization in advance havemore positive socialization outcomes (Kammeyer-Mueller & Wanberg, 2003; Wanous, 1992). Futureresearch might see how this pre-entry knowledgecontributes to newcomer adjustment by possiblyfacilitating social relationships or enhancing confi-dence needed to engage in proactive socializationbehaviors.

Practical Implications

From a management perspective, the results ofour study reinforce the importance of supervisorsand coworkers as providers of initial socialization


for organizational newcomers. Supervisors shouldbe trained to recognize how important they areas providers of feedback and support (Ostroff &Kozlowski, 1992). Training and development ses-sions for managers should emphasize that a newemployee’s first weeks on a job are likely to have aneffect on her/his subsequent adjustment and that itis within her or his supervisor’s power to createconditions to facilitate this adjustment. Supervi-sors need to take a more proactive role in theiremployees’ development, which doesn’t simplyend after two or three weeks, but remains importantduring the entire 90-day initial adjustment period.It is also necessary to train coworkers to providesupport. Studies in the past have sometimes in-cluded providing support and assistance to new-comers as elements of extra-role behavior (e.g., VanDyne & Lepine, 1998); our study suggests that or-ganizations should try to transform this supportinto a regular in-role requirement.

Our study also demonstrates that organizationsshould acknowledge that supervisor and coworkerundermining does exist and should be minimizedas much as possible because it decreases feelings ofsocial acceptance and increases withdrawal and turn-over. Research in the past has demonstrated that in-civility levels are less pronounced among employeeswho perceive that levels of fairness and equity intheir workplace are high (Blau & Andersson, 2005).Again, this appears to underscore the importance ofthe supervisors, since frontline supervisors often arethe ones who create an environment of either high orlow procedural justice.

In sum, our study of newcomer adjustment overthe first 90 days of a new job demonstrates thatinitial levels of support from coworkers and super-visors enhance hedonic tone and proactivity,whereas changes in support were related tochanges in hedonic tone and proactivity. Initiallevels of support from supervisors and coworkerswere generally positively related to work proactiv-ity, social integration, and commitment. Despitethe general trend favoring the importance of sup-port, we also found that coworker underminingdecreased social integration and withdrawal behav-iors, and supervisor undermining was uniquely re-lated to voluntary turnover.

REFERENCES

Ashford, S. J., & Black, J. S. 1996. Proactivity duringorganizational entry: The role of desire for control.Journal of Applied Psychology, 81: 199–214.

Ashford, S. J., Blatt, R., & Vande Walle, D. 2003. Reflec-tions on the looking glass: A review of research onfeedback seeking in organizations. Journal of Man-agement, 29: 773–799.

Ashforth, B. E. 2012. The role of time in socializationdynamics. In C. R. Wanberg (Ed.), The Oxford hand-book of socialization: 161–186. New York: OxfordUniversity Press.

Ashforth, B. E., Sluss, D. M., & Saks, A. M. 2007. Social-ization tactics, proactive behavior, and newcomerlearning: Integrating socialization models. Journalof Vocational Behavior, 70: 447–462.

Ayduk, O., Mendoza-Denton, R., Mischel, W., Downey,G., Peake, P. K., & Rodriguez, M. 2000. Regulatingthe interpersonal self: Strategic self-regulation forcoping with rejection sensitivity. Journal of Person-ality and Social Psychology, 79: 776–792.

Baron, R. A. 1988. Negative effects of destructive criti-cism: Impact on conflict, self-efficacy, and task per-formance. Journal of Applied Psychology, 73: 199–207.

Bauer, T. N., Bodner, T., Erdogan, B., Truxillo, D. M., &Tucker, J. S. 2007. Newcomer adjustment during or-ganizational socialization: A meta-analytic review ofantecedents, outcomes, and methods. Journal of Ap-plied Psychology, 92: 707–721.

Bauer, T. N., Erdogan, B., Liden, R. C., & Wayne, S. J.2006. A longitudinal study of the moderating role ofextraversion: Leader-member exchange, perfor-mance, and turnover during new executive develop-ment. Journal of Applied Psychology, 91: 298–310.

Bauer, T. N., & Green, S. G. 1998. Testing the combinedeffects of newcomer information seeking and man-ager behavior on socialization. Journal of AppliedPsychology, 83: 72–83.

Baumeister, R. F. 2005. The social animal. New York:Oxford University Press.

Bennett, R. J., & Robinson, S. L. 2003. The past, present,and future of workplace deviance research. In G.Greenberg (Ed.), Organizational behavior: Thestate of the science (2nd ed.): 247–281. Mahwah, NJ:Erlbaum.

Blau, G., & Andersson, L. 2005. Testing a measure ofinstigated workplace incivility. Journal of Occupa-tional and Organizational Psychology, 78: 595–614.

Bolino, M., Valcea, S., & Harvey, J. 2010. Employee,manage thyself: The potentially negative implica-tions of expecting employees to behave proactively.Journal of Occupational and Organizational Psy-chology, 83: 325–345.


Boswell, W. R., Boudreau, J. W., & Tichy, J. 2005. Therelationship between employee job change and jobsatisfaction: The honeymoon-hangover effect. Jour-nal of Applied Psychology, 90: 882–892.

Boswell, W. R., Shipp, A. J., Payne, S. C., & Culbertson,S. S. 2009. Changes in newcomer job satisfactionover time: Examining the pattern of honeymoonsand hangovers. Journal of Applied Psychology, 94:844–858.

Boudens, C. J. 2005. The story of work: A narrative anal-ysis of workplace emotion. Organization Studies,26: 1285–1306.

Bowling, N. A., & Beehr, T. A. 2006. Workplace harass-ment from the victim’s perspective: A theoreticalmodel and meta-analysis. Journal of Applied Psy-chology, 91: 988–1012.

Bradt, G., Check, J. A., & Pedraza, J. 2006. The newleaders’ 100-day action plan. Hoboken, NJ: Wiley.

Carver, C. S., & Scheier, M. F. 1998. On the self-regula-tion of behavior. New York: Cambridge University.

Chan, D. 1998. The conceptualization and analysis ofchange over time: An integrative approach incorpo-rating longitudinal mean and covariance structuresanalysis (LMACS) and multiple indicator latentgrowth modeling (MLGM). Organizational Re-search Methods, 1: 421–483.

Chan, D., & Schmitt, N. 2000. Interindividual differencesin intraindividual changes in proactivity during or-ganizational entry: A latent growth modeling ap-proach to understanding newcomer adaptation.Journal of Applied Psychology, 85: 190–210.

Chao, G. T., O’Leary-Kelly, A. M., Wolf, S., Klein, H. J., &Gardner, P. D. 1994. Organizational socialization: Itscontent and consequences. Journal of Applied Psy-chology, 79: 730–743.

Chen, G., Ployhart, R. E., Cooper-Thomas, H. C., Ander-son, N., & Bliese, P. D. 2011. The power of momen-tum: A new model of dynamic relationships betweenjob satisfaction change and turnover intentions.Academy of Management Journal, 54: 159–181.

Chiaburu, D. S., & Harrison, D. A. 2008. Do peers makethe place? Conceptual synthesis and meta-analysisof coworker effects on perceptions, attitudes, OCBs,and performance. Journal of Applied Psychology,93: 1082–1103.

Cohen, S., & Wills, T. A. 1985. Stress, social support, andthe buffering hypothesis. Psychological Bulletin, 98:310–357.

Cropanzano, R., James, K., & Konovsky, M. A. 1993.Dispositional affectivity as a predictor of work atti-

tudes and job performance. Journal of Organiza-tional Behavior, 14: 595–606.

Dougherty, T. W., Turban, D. B., & Callender, J. C. 1994.Confirming first impressions in the employment in-terview: A field study of interviewer behavior. Jour-nal of Applied Psychology, 79: 659–665.

Duffy, M. K., Ganster, D. C., & Pagon, M. 2002. Socialundermining in the workplace. Academy of Man-agement Journal, 45: 331–351.

Edmondson, A. 1999. Psychological safety and learningbehavior in work teams. Administrative ScienceQuarterly, 44: 350–383.

Elfenbein, H. A. 2008. Emotion in organizations. In A. P.Brief & J. P. Walsh (Eds.), Academy of Managementannals, vol. 1: 315–386. New York: Erlbaum.

Fisher, C. D. 1986. Organizational socialization: An inte-grative review. In G. R. Ferris & K. M. Rowland(Eds.), Research in personnel and human resourcemanagement, vol. 4: 101–145. Greenwich, CT: JAIPress.

Fisher, C. D., & Ashkanasy, N. M. 2000. The emergingrole of emotions in work life: An introduction. Jour-nal of Organizational Behavior, 21: 123–129.

Fredrickson, B. L. 2001. The role of positive emotions inpositive psychology: The broaden-and-build theoryof positive emotions. American Psychologist, 56:218–226.

Frijda, N. H. 1986. The emotions. New York: CambridgeUniversity Press.

Goldberg, L. R. 1999. A broad-bandwidth, public-do-main, personality inventory measuring the lower-level facets of several five-factor models. In I. Mer-vielde, I. J. Deary, F. De Fruyt, & F. Ostendorf (Eds.),Personality psychology in Europe, vol. 7: 7–28. Til-burg, the Netherlands: Tilburg University Press.

Grandey, A. A., Tam, A. P., & Brauburger, A. L. 2002.Affective states and traits in the workplace: Diaryand survey data from young workers. Motivationand Emotion, 26: 31–55.

Grant, A. M., & Ashford, S. J. 2008. The dynamics ofproactivity at work. In A. Brief & B. Staw (Eds.),Research in organizational behavior, vol. 28: 3–34.Bingley, UK: Emerald.

Green, D. P., Goldman, S. L., & Salovey, P. 1993. Mea-surement error masks bipolarity in affect ratings.Journal of Personality and Social Psychology, 64:1029–1041.

Griffin, M. A., Neal, A., & Parker, S. K. 2007. A newmodel of work role performance: Positive behavior


in uncertain and interdependent contexts. Academyof Management Journal, 50: 327–347.

Gruman, J. A., Saks, A. M., & Zweig, D. I. 2006. Organi-zational socialization tactics and newcomer proac-tive behaviors: An integrative study. Journal of Vo-cational Behavior, 69: 90–104.

Hanisch, K. A., & Hulin, C. L. 1991. General attitudes andorganizational withdrawal: An evaluation of a causalmodel. Journal of Vocational Behavior, 39: 110–128.

Herschcovis, M. S., & Barling, J. 2010. Toward a multi-foci approach to workplace aggression: A meta-ana-lytic review of outcomes from different perpetrators.Journal of Organizational Behavior, 31: 24–44.

Hogg, M. A., & Terry, D. J. 2000. Social identity andself-categorization processes in organizational con-texts. Academy of Management Review, 25: 121–140.

Jokisaari, M., & Nurmi, J. 2009. Change in newcomers’supervisor support and socialization outcomes afterorganizational entry. Academy of ManagementJournal, 52: 527–544.

Jones, G. R. 1986. Socialization tactics, self-efficacy, andnewcomers’ adjustments to organizations. Academyof Management Journal, 29: 262–279.

Kammeyer-Mueller, J. D., & Wanberg, C. R. 2003. Un-wrapping the organizational entry process: Disentan-gling multiple antecedents and their pathways toadjustment. Journal of Applied Psychology, 88:779–794.

Kammeyer-Mueller, J. D., Wanberg, C. R., Glomb, T. M., &Ahlburg, D. 2005. The role of temporal shifts inturnover processes: It’s about time. Journal of Ap-plied Psychology, 90: 644–658.

Kelly, J. R., & Barsade, S. G. 2001. Mood and emotions insmall groups and work teams. Organizational Be-havior and Human Decision Processes, 86: 99–130.

Lance, C. E., Vandenberg, R. J., & Self, R. M. 2000. Latentgrowth models of individual change: The case ofnewcomer adjustment. Organizational Behavior &Human Decision Processes, 83: 107–140.

Lewin, K. 1951. Field theory in social science: Selectedtheoretical papers. New York: Harper & Row.

McAllister, D. J. 1995. Affect- and cognition-based trustas foundations for interpersonal cooperation in or-ganizations. Academy of Management Journal, 38:24–59.

Miller, V. D., & Jablin, F. M. 1991. Information seekingduring organizational entry: Influences, tactics, and

a model of the process. Academy of ManagementReview, 16: 92–120.

Miner, A. G., Glomb, T. M., & Hulin, C. 2005. Experiencesampling mood and its correlates at work. Journal ofOccupational and Organizational Psychology, 78:171–193.

Moreland, R. L., & Levine, J. M. 2001. Socialization inorganizations and work groups. In M. E. Turner(Ed.), Groups at work: Theory and research: 69–112. Mahwah, NJ: Erlbaum.

Morita, J. G., Lee, T. W., & Mowday, R. T. 1993. Theregression-analog to survival analysis: A selected ap-plication to turnover research. Academy of Man-agement Journal, 36: 1430–1464.

Morrison, E. W. 1993. Longitudinal study of the effects ofinformation seeking on newcomer socialization.Journal of Applied Psychology, 78: 173–183.

Mowday, R. T., Steers, R. M., & Porter, L. W. 1979. Themeasurement of organizational commitment. Jour-nal of Vocational Behavior, 14: 224–247.

Muthén, L. K., & Muthén, B. O. 2010. MPlus user’s guide(6th ed.). Los Angeles: Muthén & Muthén.

Ng, C. F. 2006. Academics telecommuting in open anddistance education universities: Issues, challenges,and opportunities. International Review of Re-search in Open and Distance Learning, 7(2): 1–16.

Olsen, D. 1993. Work satisfaction and stress in the firstand third year of academic appointment. Journal ofHigher Education, 64: 453–471.

Ostroff, C., & Kozlowski, S. W. J. 1992. Organizationalsocialization as a learning process: The role of infor-mation acquisition. Personnel Psychology, 45: 849–874.

Parker, S. K., Williams, H. M., & Turner, N. 2006. Mod-eling the antecedents of proactive behavior at work.Journal of Applied Psychology, 91: 636–652.

Plutchik, R. 1994. The psychology and biology of emo-tion. New York: Harper Collins.

Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsa-koff, N. P. 2003. Common method biases in behav-ioral research: A critical review of the literature andrecommended remedies. Journal of Applied Psy-chology, 88: 879–903.

Rafaeli, E., Cranford, J. A., Green, A. S., Shrout, P. E., &Bolger, N. 2008. The good and bad of relationships:How social hindrance and social support affect rela-tionship feelings in daily life. Personality and So-cial Psychology Bulletin, 34: 1703–1718.


Reichers, A. E. 1987. An interactionist perspective onnewcomer socialization rates. Academy of Manage-ment Review, 12: 278–287.

Roznowski, M., & Hanisch, K. A. 1990. Building system-atic heterogeneity into work attitudes and behaviormeasures. Journal of Vocational Behavior, 36: 361–375.

Russell, J. A. 1980. A circumplex model of affect. Journalof Personality and Social Psychology, 36: 1152–1168.

Russell, J. A., & Ridgeway, D. 1983. Dimensions under-lying children’s emotion concepts. DevelopmentalPsychology, 19: 795–804.

Schein, E. H. 1978. Career dynamics: Matching individ-uals and organizational needs. Reading, MA: Ad-dison-Wesley.

Schein, E. H., & Bennis, W. 1965. Personal and organi-zational change via group methods. New York: Wi-ley.

Scollon, C., Diener, E., Oishi, S., & Biswas-Diener, R.2005. An experience sampling and cross-cultural in-vestigation of the relation between pleasant and un-pleasant affect. Cognition and Emotion, 19: 27–52.

Stallings, M., Dunham, C., Gatz, M., Baker, L., & Bengt-son, V. L. 1997. Relationships among life events andpsychological well-being: More evidence for a two-factor theory of well-being. Journal of Applied Ger-ontology, 16: 104–119.

Stohl, C. 1986. The role of memorable messages in theprocess of organizational socialization. Communica-tion Quarterly, 34: 231–249.

Tepper, B. J. 2000. Consequences of abusive supervision.Academy of Management Journal, 43: 178–190.

Tetlock, P. E. 1983. Accountability and the perseveranceof first impressions. Social Psychology Quarterly,46: 285–292.

Thoresen, C. J., Kaplan, S. A., Barsky, A. P., Warren,C. R., & de Chermont, K. 2003. The affective under-pinnings of job perceptions and attitudes: A meta-analytic review and integration. Psychological Bul-letin, 129: 914–945.

Van Dyne, L., & LePine, J. A. 1998. Helping and voiceextra-role behaviors: Evidence of construct and pre-dictive validity. Academy of Management Journal,41: 108–119.

Vinokur, A. D., Price, R. H., & Caplan, R. D. 1996. Hardtimes and hurtful partners: How financial strain af-fects depression and relationship satisfaction of un-employed persons and their spouses. Journal of Per-sonality and Social Psychology, 71: 166–179.

Vinokur, A. D., & Van Ryn, M. 1993. Social support andundermining in close relationships: Their indepen-dent effects on the mental health of unemployedpersons. Journal of Personality and Social Psychol-ogy, 65: 350–359.

Wanberg, C. R., & Kammeyer-Mueller, J. D. 2000. Predic-tors and outcomes of proactivity in the socializationprocess. Journal of Applied Psychology, 85: 373–385.

Wanous, J. P. 1992. Organizational entry: Recruitment,selection, orientation, and socialization of new-comers (2nd ed.). Reading, MA: Addison-Wesley.

Watkins, M. 2003. The first 90 days. Boston: HarvardBusiness School Press.

Watson, D. 2000. Mood and temperament. New York:Guilford Press.

Watson, D., Clark, L. A., McIntyre, C. W., & Hamaker,S. 1992. Affect, personality, and social activity. Jour-nal of Personality and Social Psychology, 63:1011–1025.

Weick, K. E. 1995. Sensemaking in organizations.Thousand Oaks, CA: Sage.

Weiss, H. M., & Beal, D. J. 2005. Reflections on effectiveevents theory. In N. M. Ashkanasy, W. J. Zerbe, &C. E. J. Härtel (Eds.), Research on emotion in organ-izations, vol. 1: 1–21. Amsterdam: Elsevier.

Weiss, H. M., & Cropanzano, R. 1996. Affective eventstheory: A theoretical discussion of the structure,causes and consequences of affective experiences atwork. In B. M. Staw, & L. L. Cummings (Eds.), Re-search in organizational behavior, vol. 18: 1–74.Greenwich., CT: JAI.

John Kammeyer-Mueller ([email protected]) is an as-sociate professor in the Department of Work and Organi-zations at the Carlson School of Management at the Uni-versity of Minnesota. He received his Ph.D. in humanresources and industrial relations from the University ofMinnesota. His research interests include personality atwork, employee socialization, the process of career de-velopment, and how work attitudes and events influencethe decision to leave a job.

Connie Wanberg ([email protected]) is the IndustrialRelations Faculty Excellence Chair in the Carlson Schoolof Management at the University of Minnesota. She re-ceived her Ph.D. in industrial/organizational psychologyfrom Iowa State University. Her research has been fo-cused on the individual experience of unemployment,job-search behavior, organizational change, employee so-cialization, and employee development.


Alex Rubenstein ([email protected]) isa doctoral candidate in the Department of Management atthe Warrington College of Business Administration at theUniversity of Florida. His research has been focused onnewcomer socialization and adjustment, employee turn-over, emotions/mood, and individual differences in in-telligence and personality.

Zhaoli Song ([email protected]) is an associate profes-sor in the Department of Management and Organization

at the NUS School of Business at the National Universityof Singapore. He received his Ph.D. in human resourcesand industrial relations from the University of Minne-sota. His research interests include behavior genetics, jobsearch and unemployment, momentary emotional expe-rience, work-family boundaries, Chinese management,and research design and methods.


Copyright of Academy of Management Journal is the property of Academy of Managementand its content may not be copied or emailed to multiple sites or posted to a listserv withoutthe copyright holder's express written permission. However, users may print, download, oremail articles for individual use.

Documents

SUPPORT, UNDERMINING, AND NEWCOMER ......make the process of adjustment especially difficult. Another unresolved issue for socialization re-search involves time. Socialization is a